Abstract
We assume that visual object recognition occurs by mapping incoming visual data onto a set of object categories. A claim is made that the world is structured in such a way that allows a definition of natural classes and that these classes should be the basis for visual categorization. Therefore, a vision system that is to perform successful visual categorization in the natural world must be able to recover these natural classes.
© 1986 Optical Society of America
PDF ArticleMore Like This
Philippe G. Schyns and Heinrich H. Bulthoff
ThVV4 OSA Annual Meeting (FIO) 1992
Ronald Wu and Henry Stark
ThD2 Signal Recovery and Synthesis (SRS) 1986
Gerd Häusler
CTh2A.1 Computational Optical Sensing and Imaging (COSI) 2019